Back to all rerankers

Zerank 1 Small

Compact 1.7B parameter model with novel ELO-based training on synthetic pairwise data for calibrated relevance scores. Half the cost of competing rerankers while maintaining strong ranking quality. If you want to compare the best rerankers for your data, try Agentset.

Leaderboard Rank
#5
of 12
ELO Rating
1539
#5
Win Rate
55.0%
#5
Accuracy (nDCG@10)
0.083
#8
Latency
248ms
#1

Model Information

Provider
ZeroEntropy
License
Open Source
Price per 1M tokens
$0.025
Release Date
2025-07-10
Model Name
zerank-1-small
Total Evaluations
3300

Performance Record

Wins1816 (55.0%)
Losses1355 (41.1%)
Ties129 (3.9%)
Wins
Losses
Ties

Rerankers Are Just One Piece of RAG

Agentset gives you a managed RAG pipeline with the top-ranked models and best practices baked in. No infrastructure to maintain, no reranking to configure.

Trusted by teams building production RAG applications

5M+
Documents
1,500+
Teams
99.9%
Uptime

Performance Overview

ELO ratings by dataset

Zerank 1 Small's ELO performance varies across different benchmark datasets, showing its strengths in specific domains.

Zerank 1 Small - ELO by Dataset

Detailed Metrics

Dataset breakdown

Performance metrics across different benchmark datasets, including accuracy and latency percentiles.

PG

ELO 161858.0% WR319W-231L-0T

Accuracy Metrics

nDCG@5
0.000
nDCG@10
0.000
Recall@5
0.000
Recall@10
0.000

Latency Distribution

Mean
287ms
P50 (Median)
274ms
P90
345ms

DBPedia

ELO 157359.3% WR326W-175L-49T

Accuracy Metrics

nDCG@5
0.000
nDCG@10
0.000
Recall@5
0.000
Recall@10
0.000

Latency Distribution

Mean
226ms
P50 (Median)
223ms
P90
241ms

business reports

ELO 154753.1% WR292W-249L-9T

Accuracy Metrics

nDCG@5
0.000
nDCG@10
0.000
Recall@5
0.000
Recall@10
0.000

Latency Distribution

Mean
251ms
P50 (Median)
242ms
P90
298ms

MSMARCO

ELO 153548.7% WR268W-240L-42T

Accuracy Metrics

nDCG@5
0.000
nDCG@10
0.000
Recall@5
0.000
Recall@10
0.000

Latency Distribution

Mean
217ms
P50 (Median)
214ms
P90
227ms

arguana

ELO 150754.7% WR301W-239L-10T

Accuracy Metrics

nDCG@5
0.279
nDCG@10
0.375
Recall@5
0.520
Recall@10
0.820

Latency Distribution

Mean
256ms
P50 (Median)
254ms
P90
280ms

FiQa

ELO 145256.4% WR310W-221L-19T

Accuracy Metrics

nDCG@5
0.114
nDCG@10
0.124
Recall@5
0.098
Recall@10
0.125

Latency Distribution

Mean
248ms
P50 (Median)
244ms
P90
266ms

Build RAG in Minutes, Not Months

Agentset gives you a complete RAG API with top-ranked rerankers and embedding models built in. Upload your data, call the API, and get accurate results from day one.

import { Agentset } from "agentset";

const agentset = new Agentset();
const ns = agentset.namespace("ns_1234");

const results = await ns.search(
  "What is multi-head attention?"
);

for (const result of results) {
  console.log(result.text);
}

Compare Models

See how it stacks up

Compare Zerank 1 Small with other top rerankers to understand the differences in performance, accuracy, and latency.